• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 62 subscribers
  • Views 195 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM3409: Testing constant current source without LEDs

Ulrich Scherhag
Prodigy 10 points
Part Number: LM3409
Other Parts Discussed in Thread: CSD25404Q3

Dear team,

I am designing a CC-source for high speed pulsed LED/VCSEL applications using the LM3409. During my design experiments I keep killing the illumination by error. How can I test the circuit without illumination?

Best Regards

Ulrich

  • 0
    TI__Mastermind 28655 points

    Hello Ulrich.

    What current are you running?  What is the output voltage?  There are a few things you can do such as string power diodes in series to get close to the voltage you are running.  Another possibility is a load resistor however you have to be careful to make sure that the LM3409 can start if using a resistor.  The LM3409 requires the output to be above 1.243V to charge the off-time capacitor, if it cannot get there from the initial first on-time it will go into maximum off-time and not start.  LED loads don't do this since they only start conduct current when near operating voltage.

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Prodigy 10 points

    Hello Irwin,

    thank you for the reply.

    I am testing multiple LEDs and VSELS, thus current and output voltage are varying. The first approach is:

    V_in 12V

    V_out 9.6V

    I_LED 0,5A

    Following the recommendations and examples from the datasheet I aim at a switching frequency of 500kHz. This leads to:

    C_off 560pF

    R_off 3.9kOhm

    real f_sw 533kHz

    L 5.6uH

    that assumption correct

    Further I use R_IADJ to limit I_LED:

    R_shunt 75mOhm

    R_IADJ 82kOhm

    As P-FET I am using CSD25404Q3.

    Regarding your recommendations: If I am using a load resistor, the resistor itself will limit the current depending on the voltage, or in other words: the output voltage depends on the current. This will influence the whole circuit, e.g. switching frequency. Is that assumption correct?

    So the string diodes should provide a more realistic result?

    Best regards

    Ulrich

  • 0 in reply to Ulrich Scherhag
    TI__Mastermind 28655 points

    Hi Ulrich,

    Using a resistor will not closely resemble an LED load as you say, the voltage will vary with current.  It also has another drawback if the output is close to the input voltage and high ripple current.  The LM3409 regulates peak current, if it cannot reach peak current the MOSFET will just stay on.  Note this will not damage anything since the current is limited by the load.

    A diode string operates similar to an LED.  I've designed a lot of LED loads using diodes.  Try the higher Vf diodes like RS1J, RS1K, RS1M, or through hole versions such as the 1N4007.  You can use larger parts if thermal variation is an issue.  Stack them in series and use a lab supply to get close to the load voltage you are trying to match.

    I'm not sure why you picked such a low current sense resistor value.  Regulation will be more accurate if you have a higher Viadj.  This design will also exhibit Datasheet 8.3.4 and Figure 24 on page 15.  Your design will have 4 mV of Vcs ripple (Viadj - Viadjval) and the regulation will not be accurate.  For the best accuracy I would using Rcs at 0.45 ohm (or a little higher if you want to be at 0.5A average) and remove Riadj (82 Kohm).  Current ripple is low so the peak and current sense valley will be right around 24 mV now.

    Your design calculations (I did not include efficiency in the calculations):

    Vin 12 V
    Vout 9.5 V
    Iled 0.5 A
    Coff 5.6E-10 F
    Roff 3900 ohm
    Lind 0.000056 H
    Rcs 0.075 ohm
    Riadj 82000 ohm
         
    Voff 1.243 V
    Vcs 0.2486 V
    Iiadj 0.0000005 A
         
    toff 3.17201E-07 s
    duty 0.791666667  
    ton 1.20537E-06 sec
    Fsw 656785.8121 Hz
    dioff 0.05381094 Apkpk
    dion 0.05381094 Apkpk
    Viadj 0.041 V
    Ipkmax 3.314666667 Apk
    Ipk 0.546666667 Apk
    Iave 0.519761197 Aave
    Viadjval 0.03696418 V

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Prodigy 10 points

    Hi Irwin,

    Thanks for the detailed answer.
    In the meantime I have an electronic load (BK Precision 8602) available to test the circuit. Any concerns regarding simulating the Diodes with a constant voltage load?

    I chose the small resistor as the circuit should also be able to deliver currents up to 4A with other VCSELs. In retrospect, of course, it sounds more reasonable to adjust the current measuring resistor instead of Riadj. Thanks for the hint.

    Best Regards
    Ulrich

  • 0 in reply to Ulrich Scherhag
    TI__Mastermind 28655 points

    Hello Ulrich,

    An electronic load should be fine, just note that you may run into a start-up issue with an electronic load.  If it creates enough load when the output starts to rise and the LM3409 output cannot get above 1.243V it will go into maximum off-time and may not start.  You will be able to tell this when testing.

    Best Regards,